Everything is a number in C, but binary choices are very common in logic. It's at the core of every 'if' statement: having to re-invent booleans in so many different forms for different uses is just painful.
If Lisp were that good, it would have done better in the marketplace. The 'layers of abstraction' of Lisp, and its programmers focus on ignoring what is going on at lower levels, waving a lot of parentheses shaped magic wands and saying 'and then a miracle occurs', and worse yet using recursion to call the same miracle again, and again, and again without paying attention to what that miracle costs, led to terrible programming practices. And the garbage collecion problem keeps turning up, and at my last investigation has no solution. If you've abstracted away all the knowledge of lower level processes, you've surrendered your available tools to mitigate the issues.
Yes, I've worked with Lisp and some of its variants. Unless it's advanced a lot and lost its fascination with recursion for its own sake, it's never going to scale well and will always underperform.
No, there are plenty of active forums. But people have to contribute to them to make them rich in good answers, and alternative approaches. And the Weblogs for specific projects, or bug reporting tools like Bugzilla hosted at Sourceforge, provide a lot of the service that Usenet formerly did, so Usenet is seriously reduced.
It would only be justified if it worked, and you could be sure that they weren't just bringing in or fostering their own set of murderous wild-eyed lunatics.
Have you looked at Iraq and Afghanistan lately? The Taleban is on the rebound, civil services have gone back to the same levels they were under the _Russians_ in Afghanistan, and Iraq has become a recruiting ground for Muslim terrorists worldwide. The invasion has, in fact, made it less safe than it was under Hussein, at least for those in Baghdad.
By putting such a hearing at the end of a lame duck presidency, and an extremely corporate friendly one at that, they're guaranteeing jobs for the current set of bureaucrats for at least another six months, even if the new president thinks they're all worthless and should be canned. They're looking for new work to do in the new adminstration, and expanding their existing bureaucracy to 'organize' the existing mess.
If they had a chance of discarding the existing software patent and current ludicrous copyright laws and setting out a sane standard, with clear standards for fair use and clear new patent guidelines that dump the current use of patent libraries to frighten developers away from releasing their smaller ideas, I'd say it would be great. But it's unlikely to occur in Mr. Barack's first term: he's going to have to set new standards in bureaucratic policy, as well. And that would take people at the hearings such as Richard Stallman as the author of the GPL.
No, it's not that expensive if you're not doing it in general. And it seems to only be a 'legal risk' if you'e not doing it for the the law enforcers. Then the legal risk goes to nil.
The difference is especially artificial in the RedHat world: the difference is primarily support levels and licensing for some fascinating tools that the average installation does not need, such as number of CPU's, number of virtual installations, and some very sophisticated clustering tools.
Not all of us, thank you. I'm a strong proponent of modifying the existing deployed toolkits and communicating with the authors to integrate your features or changes. This often works quite well, and the exceptions seem to be people who refuse to use the GPL. (Dan Bernstein and qumail come to mind, although he has since changed his licensing model.)
Their policy is not to 'make sense'. Their policy is to frighten people. An insane attack dog is more frightening, and in a number of more ways more effective, than a well-trained guard dog to keep people off the territory where you let the dog loose, even if you do not in fact own that territory and have no legal cause to let that dog hurt anyone.
Plenty of people in the music industry, especially in production and distribution, have mastered this art for many years, against agents, performers, and normal purchasers. This is just another form of the 'trial by champions' or effectively 'trial by mercenaries in suits' that legal systems have provided since the time of the crucifixion of Jesus and Pontius Pilate washing his hands of the mess.
MRI chambers due to the poor interactions if high magnetic fields and a lot of computer equipment comes to mind. Other radiological work has similar issues: biological work where it's easier to destroy paper than to sterilize computer also comes to mind.
Copyright no longer effectively expires, thanks to Disney Corporation protecting Mickey Mouse. To quote from a writeup of the current US law, "the Act extended these terms to life of the author plus 70 years and for works of corporate authorship to 120 years after creation or 95 years after publication, whichever endpoint is earlier."
I was referring, not to your magazines, but to others writing game guides and publishing their own work on the web where it is freely readable or downloadable. And I thought you were referring to the copyrights of the game developers, not of your magazines' publishers.
Please excuse my confusion: we've seen so many cases of game and software developers trying to control publication of game guides and walkthroughs that I thought you meant those.
> downloaded pictures of copyrighted materials, is a crime in some countries.
Not necessarily. The 'fair use' exceptions, while not as clearly defined as a programmer might like, certainly exist and allow 'criticism' of copyrighted work. I think that careful use of screenshots would be fair use, especially if you're careful not to violate the trademarks and claim that your guide is official material from the game designers.
Now we have Google and Wikipedia to find on-line gaming guides. I don't think you need worry about this lack.
Some of the X-com and Thief games have shown up on Steam, the PC gaming service that includes Half-Life and Darwinia. I'm thrilled: now I don't have to hang onto my old media, and I can play it on any PC I have access to.
Not everyone has/bin before/usr/bin. In particular, many folks have/usr/local/bin before either, and emacs is often installed there in UNIX systems that don't come with Emacs (though these are getting rarer).
The point is that you cannot just assume that a binary that you didn't insall and you didn't package manage is in a particular location on someone else's system. In addition, even if emacs is linked to in/bin/emacs, the next update of the emacs package is likely to replace it and put it right back where you tried to hide it from.
What you are doing is exactly what NVidia did to the OpenGL libraries in their Linux installer, and it's a source of real package management problems. It's what someone does on their home environment, and leaves as a booby trap for others to cope with. If you don't want emacs, don't install it, or make changes that affect your personal environment, not everyone else's. For example, having $HOME/bin early in your PATH allows you to put a script there called 'emacs' that says "don't use this!", but doesn't step on package management or other people's use of the server.
Depends on the server. For a hosted server, the client and the local admin need access. For a development server, the developers and the admin need access. For a personal box on the network, owned by someone comfortable with their own amdinistration, I still want root access as the admin so I can do critical updates when they're not available or arrange changes in backups at need. In a large deployed environment, there may be half a dozen admins with different shifts and days on call that need access.
Few people use these commands fully. The ability to '-prune -xdev -o' under find, to prevent it descending mounted directories, is wonderful. The 'grep --exclude' option is also underused.
For advanced students, I suggest a 2-week course on the intricacies of sed and awk. With good mastery of those tools, much of the desire for Perl evaporates.
Not by me! Sharing root passwords was _nasty_, and sudo provided much better logging of who was going in as root and forcing them to use their own passwords, not a root password.
This is not the venue for teaching. I use patch submissions for new projects I like, and education in the workplace about already existing tools, and activity on developer discussion groups.
RPM has gotten pretty good about reporting dynamic libraries and recording their dependencies. It's when different packages need different libraries, and are built from different sources or distributions without ever being RPM managed, that life gets nutty. If I see one more package that is handbuilt from both Apache 1.3 modules and the latest untested CPAN tools, and have to unweave the dependencies and backport libraries for it, well, I'm going to be unhappy about it.
Slashdot Mirror
Everything is a number in C, but binary choices are very common in logic. It's at the core of every 'if' statement: having to re-invent booleans in so many different forms for different uses is just painful.
If Lisp were that good, it would have done better in the marketplace. The 'layers of abstraction' of Lisp, and its programmers focus on ignoring what is going on at lower levels, waving a lot of parentheses shaped magic wands and saying 'and then a miracle occurs', and worse yet using recursion to call the same miracle again, and again, and again without paying attention to what that miracle costs, led to terrible programming practices. And the garbage collecion problem keeps turning up, and at my last investigation has no solution. If you've abstracted away all the knowledge of lower level processes, you've surrendered your available tools to mitigate the issues.
Yes, I've worked with Lisp and some of its variants. Unless it's advanced a lot and lost its fascination with recursion for its own sake, it's never going to scale well and will always underperform.
No, there are plenty of active forums. But people have to contribute to them to make them rich in good answers, and alternative approaches. And the Weblogs for specific projects, or bug reporting tools like Bugzilla hosted at Sourceforge, provide a lot of the service that Usenet formerly did, so Usenet is seriously reduced.
It would only be justified if it worked, and you could be sure that they weren't just bringing in or fostering their own set of murderous wild-eyed lunatics.
Have you looked at Iraq and Afghanistan lately? The Taleban is on the rebound, civil services have gone back to the same levels they were under the _Russians_ in Afghanistan, and Iraq has become a recruiting ground for Muslim terrorists worldwide. The invasion has, in fact, made it less safe than it was under Hussein, at least for those in Baghdad.
By putting such a hearing at the end of a lame duck presidency, and an extremely corporate friendly one at that, they're guaranteeing jobs for the current set of bureaucrats for at least another six months, even if the new president thinks they're all worthless and should be canned. They're looking for new work to do in the new adminstration, and expanding their existing bureaucracy to 'organize' the existing mess.
If they had a chance of discarding the existing software patent and current ludicrous copyright laws and setting out a sane standard, with clear standards for fair use and clear new patent guidelines that dump the current use of patent libraries to frighten developers away from releasing their smaller ideas, I'd say it would be great. But it's unlikely to occur in Mr. Barack's first term: he's going to have to set new standards in bureaucratic policy, as well. And that would take people at the hearings such as Richard Stallman as the author of the GPL.
No, it's not that expensive if you're not doing it in general. And it seems to only be a 'legal risk' if you'e not doing it for the the law enforcers. Then the legal risk goes to nil.
The difference is especially artificial in the RedHat world: the difference is primarily support levels and licensing for some fascinating tools that the average installation does not need, such as number of CPU's, number of virtual installations, and some very sophisticated clustering tools.
Not all of us, thank you. I'm a strong proponent of modifying the existing deployed toolkits and communicating with the authors to integrate your features or changes. This often works quite well, and the exceptions seem to be people who refuse to use the GPL. (Dan Bernstein and qumail come to mind, although he has since changed his licensing model.)
No, only that they're incompoetent in losing really important objects and concealing that it was their fault.
Their policy is not to 'make sense'. Their policy is to frighten people. An insane attack dog is more frightening, and in a number of more ways more effective, than a well-trained guard dog to keep people off the territory where you let the dog loose, even if you do not in fact own that territory and have no legal cause to let that dog hurt anyone.
Plenty of people in the music industry, especially in production and distribution, have mastered this art for many years, against agents, performers, and normal purchasers. This is just another form of the 'trial by champions' or effectively 'trial by mercenaries in suits' that legal systems have provided since the time of the crucifixion of Jesus and Pontius Pilate washing his hands of the mess.
MRI chambers due to the poor interactions if high magnetic fields and a lot of computer equipment comes to mind. Other radiological work has similar issues: biological work where it's easier to destroy paper than to sterilize computer also comes to mind.
Copyright no longer effectively expires, thanks to Disney Corporation protecting Mickey Mouse. To quote from a writeup of the current US law, "the Act extended these terms to life of the author plus 70 years and for works of corporate authorship to 120 years after creation or 95 years after publication, whichever endpoint is earlier."
I was referring, not to your magazines, but to others writing game guides and publishing their own work on the web where it is freely readable or downloadable. And I thought you were referring to the copyrights of the game developers, not of your magazines' publishers.
Please excuse my confusion: we've seen so many cases of game and software developers trying to control publication of game guides and walkthroughs that I thought you meant those.
You wrote:
> downloaded pictures of copyrighted materials, is a crime in some countries.
Not necessarily. The 'fair use' exceptions, while not as clearly defined as a programmer might like, certainly exist and allow 'criticism' of copyrighted work. I think that careful use of screenshots would be fair use, especially if you're careful not to violate the trademarks and claim that your guide is official material from the game designers.
Now we have Google and Wikipedia to find on-line gaming guides. I don't think you need worry about this lack.
Some of the X-com and Thief games have shown up on Steam, the PC gaming service that includes Half-Life and Darwinia. I'm thrilled: now I don't have to hang onto my old media, and I can play it on any PC I have access to.
So the Iraqi Parliament is a bunch of line cutters? That would explain a lot.
I'm afraid not. The AT&T style UNIX's didn't support the '-e' option for the 'echo' command. Use printf instead.
Not everyone has /bin before /usr/bin. In particular, many folks have /usr/local/bin before either, and emacs is often installed there in UNIX systems that don't come with Emacs (though these are getting rarer).
The point is that you cannot just assume that a binary that you didn't insall and you didn't package manage is in a particular location on someone else's system. In addition, even if emacs is linked to in /bin/emacs, the next update of the emacs package is likely to replace it and put it right back where you tried to hide it from.
What you are doing is exactly what NVidia did to the OpenGL libraries in their Linux installer, and it's a source of real package management problems. It's what someone does on their home environment, and leaves as a booby trap for others to cope with. If you don't want emacs, don't install it, or make changes that affect your personal environment, not everyone else's. For example, having $HOME/bin early in your PATH allows you to put a script there called 'emacs' that says "don't use this!", but doesn't step on package management or other people's use of the server.
And, of course, emacs is normally in /usr/bin, which is for larger user utilities. (Some systems symlink /bin to /usr/bin, but certainly not all)
But try running 'emacs -e psychonalyze-pinhead', or 'emacs -e tower-of-hanoi' if you have a full GNU emacs installation.
Depends on the server. For a hosted server, the client and the local admin need access. For a development server, the developers and the admin need access. For a personal box on the network, owned by someone comfortable with their own amdinistration, I still want root access as the admin so I can do critical updates when they're not available or arrange changes in backups at need. In a large deployed environment, there may be half a dozen admins with different shifts and days on call that need access.
Even your 'admin and their ackup' are two people.
I prefer:
du --max-depth=1 | sort -n
This handles hardlinks among the directory better, and will also report any .files or .directories hidden at the top level.
Paying attention to dotfiles is very valuable.
Few people use these commands fully. The ability to '-prune -xdev -o' under find, to prevent it descending mounted directories, is wonderful. The 'grep --exclude' option is also underused.
For advanced students, I suggest a 2-week course on the intricacies of sed and awk. With good mastery of those tools, much of the desire for Perl evaporates.
Not by me! Sharing root passwords was _nasty_, and sudo provided much better logging of who was going in as root and forcing them to use their own passwords, not a root password.
This is not the venue for teaching. I use patch submissions for new projects I like, and education in the workplace about already existing tools, and activity on developer discussion groups.
RPM has gotten pretty good about reporting dynamic libraries and recording their dependencies. It's when different packages need different libraries, and are built from different sources or distributions without ever being RPM managed, that life gets nutty. If I see one more package that is handbuilt from both Apache 1.3 modules and the latest untested CPAN tools, and have to unweave the dependencies and backport libraries for it, well, I'm going to be unhappy about it.
Except when it doesn't. That 'secret sauce' is powerful stuff, but Debian has a rather different approach to package management.
Now try getting the same set of libraries twice running on a gentoo box, as the compiler changes behind your back.